m5-forecasting-accuracy-simplest-model.ipynb

{"cells":[{"metadata":{},"cell_type":"markdown","source":"# M5 Competition\n\nThe M5 comptition ran from 2 March to 30 June 2020. Basis of the competition is to predict sales forecasts for walm,art stores. For that, we use hierarchical sales data, generously made available by Walmart, starting at the item level and aggregating to that of departments, product categories, stores in three geographical areas of the US: California, Texas, and Wisconsin.\n\nEach row contains an id that is a concatenation of an item_id and a store_id, which is either validation (corresponding to the Public leaderboard), or evaluation (corresponding to the Private leaderboard). \n\nWe are predicting 28 forecast days (F1-F28) of items sold for each row. For the **validation rows**, this corresponds to d_1914 - d_1941, and for the **evaluation rows**, this corresponds to d_1942 - d_1969. (Note: a month before the competition close, the ground truth for the validation rows will be provided.)\n\nDetailed Information can be found [here](https://mofc.unic.ac.cy/m5-competition/) and [here](https://www.kaggle.com/c/m5-forecasting-accuracy/data).\n\nAn overview of the data given, can be seen here:"},{"metadata":{},"cell_type":"markdown","source":"Data exists in three files:\n1. File 1: “calendar.csv”: Contains information about the dates the products are sold.\n    * date: The date in a “y-m-d” format.\n    * wm_yr_wk: The id of the week the date belongs to.\n    * weekday: The type of the day (Saturday, Sunday, …, Friday).\n    * wday: The id of the weekday, starting from Saturday.\n    * month: The month of the date.\n    * year: The year of the date.\n    * event_name_1: If the date includes an event, the name of this event.\n    * event_type_1: If the date includes an event, the type of this event.\n    * event_name_2: If the date includes a second event, the name of this event.\n    * event_type_2: If the date includes a second event, the type of this event.\n    * snap_CA, snap_TX, and snap_WI: A binary variable (0 or 1) indicating whether the stores of CA, TX or WI allow SNAP  purchases on the examined date. 1 indicates that SNAP purchases are allowed.\n\n\n2. File 2: “sell_prices.csv”: Contains information about the price of the products sold per store and date.\n    * store_id: The id of the store where the product is sold. \n    * item_id: The id of the product.\n    * wm_yr_wk: The id of the week.\n    * sell_price: The price of the product for the given week/store. The price is provided per week (average across seven days). If not available, this means that the product was not sold during the examined week. Note that although prices are constant at weekly basis, they may change through time (both training and test set).  \n\n\n3. File 3: “sales_train.csv”: Contains the historical daily unit sales data per product and store.\n    * item_id: The id of the product.\n    * dept_id: The id of the department the product belongs to.\n    * cat_id: The id of the category the product belongs to.\n    * store_id: The id of the store where the product is sold.\n    * state_id: The State where the store is located.\n    * d_1, d_2, …, d_i, … d_1941: The number of units sold at day i, starting from 2011-01-29. \n"},{"metadata":{},"cell_type":"markdown","source":"![grafik.png](attachment:grafik.png)","attachments":{"grafik.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"Aknowledgements\nAs a starting point I mainly used two notebooks:  \nhttps://www.kaggle.com/bountyhunters/baseline-lstm-with-keras-0-8#Future-Improvements  \nhttps://www.kaggle.com/kuberiitb/let-s-start-from-here-beginners-data-analysis#Data-Cleaning  \nhttps://www.kaggle.com/kyakovlev/m5-custom-features"},{"metadata":{},"cell_type":"markdown","source":"# Importing Modules and functions"},{"metadata":{"trusted":true},"cell_type":"code","source":"timesteps = 28\nstartDay = 0","execution_count":null,"outputs":[]},{"metadata":{"_cell_guid":"b1076dfc-b9ad-4769-8c92-a6c4dae69d19","_uuid":"8f2839f25d086af736a60e9eeb907d3b93b6e0e5","trusted":true},"cell_type":"code","source":"import pandas as pd\nimport numpy as np\nimport sklearn as skl\nimport matplotlib.pyplot as plt\nimport scipy as sc\nimport gc #importing garbage collector\nimport time\nimport sys\nfrom scipy import signal\nfrom itertools import chain\n\n\nimport warnings\nwarnings.filterwarnings('ignore')\n\n%matplotlib inline  \n\nSEED = 42\n#Pandas - Displaying more rorws and columns\npd.set_option(\"display.max_rows\", 500)\npd.set_option(\"display.max_columns\", 500)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"def reduce_mem_usage(df, verbose=True):\n    numerics = ['int8','int16', 'int32', 'int64', 'float16', 'float32', 'float64']\n    start_mem = df.memory_usage().sum() / 1024**2    \n    for col in df.columns:\n        col_type = df[col].dtypes\n        if col_type in numerics:\n            c_min = df[col].min()\n            c_max = df[col].max()\n            if str(col_type)[:3] == 'int':\n                if c_min > np.iinfo(np.int8).min and c_max < np.iinfo(np.int8).max:\n                    df[col] = df[col].astype(np.int8)\n                elif c_min > np.iinfo(np.int16).min and c_max < np.iinfo(np.int16).max:\n                    df[col] = df[col].astype(np.int16)\n                elif c_min > np.iinfo(np.int32).min and c_max < np.iinfo(np.int32).max:\n                    df[col] = df[col].astype(np.int32)\n                elif c_min > np.iinfo(np.int64).min and c_max < np.iinfo(np.int64).max:\n                    df[col] = df[col].astype(np.int64)  \n            else:\n                if c_min > np.finfo(np.float16).min and c_max < np.finfo(np.float16).max:\n                    df[col] = df[col].astype(np.float16)\n                elif c_min > np.finfo(np.float32).min and c_max < np.finfo(np.float32).max:\n                    df[col] = df[col].astype(np.float32)\n                else:\n                    df[col] = df[col].astype(np.float64)    \n    end_mem = df.memory_usage().sum() / 1024**2\n    if verbose: print('Mem. usage decreased to {:5.2f} Mb ({:.1f}% reduction)'.format(end_mem, 100 * (start_mem - end_mem) / start_mem))\n    return df","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Data Overview"},{"metadata":{"trusted":true},"cell_type":"code","source":"#!kaggle competitions download -c m5-forecasting-accuracy","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_train = pd.read_csv('../input/m5-forecasting-accuracy/sales_train_evaluation.csv')\n#df_prices = pd.read_csv('../input/m5-forecasting-accuracy/sell_prices.csv')\ndf_days = pd.read_csv('../input/m5-forecasting-accuracy/calendar.csv')\n\ndf_train = reduce_mem_usage(df_train)\n#df_prices = reduce_mem_usage(df_prices)\ndf_days = reduce_mem_usage(df_days)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"As we want to have each day as row and 30490 items' sales as columns (features) we take the transpose of the dataframe"},{"metadata":{"trusted":true},"cell_type":"code","source":"df_train=df_train.T","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_train.iloc[:10,:10]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_train.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days.tail(10)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days.shape","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#my_df_summary(df_train)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#my_df_summary(df_prices)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Feature Engineering"},{"metadata":{},"cell_type":"markdown","source":"The feature wday has the numerical weekdays in it. We can create a new feature to separate between weekdays (Monday to Friday) and Weekend (Saturday or Sunday). Weekdays equal numbers wday > 2."},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days['is_workday'] = 0\ndf_days['is_workday'].loc[df_days['wday']>2] =1\ndf_days['is_workday'] = df_days['is_workday'].astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days['is_event_day'] = [1 if x ==False else 0 for x in df_days['event_name_1'].isnull()] \ndf_days['is_event_day'] = df_days['is_event_day'].astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days[\"date\"] = pd.to_datetime(df_days['date'])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days['week'] = df_days[\"date\"].dt.week\ndf_days['week'] = df_days['week'].astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days['num_events_week'] = df_days.groupby(by=['year','week'])['is_event_day'].transform('sum')\ndf_days['num_events_week'] = df_days['num_events_week'].astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days['is_event_week'] = [1 if x >0 else 0 for x in df_days['num_events_week']]\ndf_days['is_event_week'] = df_days['is_event_week'].astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days.info()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days.set_index('date', inplace=True)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"day_after_event = df_days[df_days['is_event_day']==1].index.shift(1,freq='D')\ndf_days['is_event_day_after'] = 0\ndf_days['is_event_day_after'][df_days.index.isin(day_after_event)] = 1\ndf_days['is_event_day_after'] = df_days['is_event_day_after'].astype(np.int8)\n\ndel day_after_event","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"day_before_event = df_days[df_days['is_event_day']==1].index.shift(-1,freq='D')\ndf_days['is_event_day_before'] = 0\ndf_days['is_event_day_before'][df_days.index.isin(day_before_event)] = 1\ndf_days['is_event_day_before'] = df_days['is_event_day_before'].astype(np.int8)\n\ndel day_before_event","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days.loc[:, \"is_sport_event\"] = ((df_days[\"event_type_1\"] == \"Sporting\") | (df_days[\"event_type_2\"] == \"Sporting\")).astype(\"int8\")\ndf_days.loc[:, \"is_cultural_event\"] = ((df_days[\"event_type_1\"] == \"Cultural\") | (df_days[\"event_type_2\"] == \"Cultural\")).astype(\"int8\")\ndf_days.loc[:, \"is_national_event\"] = ((df_days[\"event_type_1\"] == \"National\") | (df_days[\"event_type_2\"] == \"National\")).astype(\"int8\")\ndf_days.loc[:, \"is_religious_event\"] = ((df_days[\"event_type_1\"] == \"Religious\") | (df_days[\"event_type_2\"] == \"Religious\")).astype(\"int8\")","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# adding 'id' column as well as 'cat_id', 'dept_id' and 'state_id', then changing the type to 'categorical'\n#df_prices.loc[:, \"id\"] = df_prices.loc[:, \"item_id\"] + \"_\" + df_prices.loc[:, \"store_id\"] + \"_validation\"\n#df_prices['state_id'] = df_prices['store_id'].str.split('_',expand=True)[0]\n#df_prices = pd.concat([df_prices, df_prices[\"item_id\"].str.split(\"_\", expand=True)], axis=1)\n#df_prices = df_prices.rename(columns={0:\"cat_id\", 1:\"dept_id\"})\n#df_prices[[\"store_id\", \"item_id\", \"cat_id\", \"dept_id\", 'state_id']] = df_prices[[\"store_id\",\"item_id\", \"cat_id\", \"dept_id\", 'state_id']].astype(\"category\")\n#df_prices = df_prices.drop(columns=2)\n#\n## It seems that leadings zero values in each train_df item row are not real 0 sales but mean absence for the item in the store\n## we can safe some memory by removing such zeros\n#release_df = df_prices.groupby(['store_id','item_id'])['wm_yr_wk'].agg(['min']).reset_index()\n#release_df.columns = ['store_id','item_id','release']\n#\n#df_prices = pd.merge(df_prices, release_df, on=['store_id','item_id'])\n#del release_df","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# We can do some basic aggregations\n#df_prices['price_max'] = df_prices.groupby(['store_id','item_id'])['sell_price'].transform('max')\n#df_prices['price_min'] = df_prices.groupby(['store_id','item_id'])['sell_price'].transform('min')\n#df_prices['price_std'] = df_prices.groupby(['store_id','item_id'])['sell_price'].transform('std')\n#df_prices['price_mean'] = df_prices.groupby(['store_id','item_id'])['sell_price'].transform('mean')\n#df_prices['price_norm'] = df_prices['sell_price']/df_prices['price_max']\n#\n#df_prices['price_nunique'] = df_prices.groupby(['store_id','item_id'])['sell_price'].transform('nunique')\n#df_prices['item_nunique'] = df_prices.groupby(['store_id','sell_price'])['item_id'].transform('nunique').astype(np.int8)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_days[['wm_yr_wk','month','year']].head()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# I would like some \"rolling\" aggregations\n# but would like months and years as \"window\"\n#calendar_prices = df_days[['wm_yr_wk','month','year']]\n#calendar_prices = calendar_prices.drop_duplicates(subset=['wm_yr_wk'])\n#df_prices = df_prices.merge(calendar_prices[['wm_yr_wk','month','year']], on=['wm_yr_wk'], how='left')\n#del calendar_prices","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#df_prices['price_momentum'] = df_prices['sell_price']/df_prices.groupby(['store_id','item_id'])['sell_price'].transform(lambda x: x.shift(1))\n#df_prices['price_momentum_m'] = df_prices['sell_price']/df_prices.groupby(['store_id','item_id','month'])['sell_price'].transform('mean')\n#df_prices['price_momentum_y'] = df_prices['sell_price']/df_prices.groupby(['store_id','item_id','year'])['sell_price'].transform('mean')","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#df_prices","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#df_prices['release'] = df_prices['release'] - df_prices['release'].min()\n#df_prices['release'] = df_prices['release'].astype(np.int16)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"gc.collect()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_train_full = df_train.copy()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Generating Train and Test Data Option a: Limited features"},{"metadata":{},"cell_type":"markdown","source":"In the next part we have to decide how many features we want to take for test and training datasets. In the first part we are taking only one extra feature (i.e. limited features)."},{"metadata":{"trusted":true},"cell_type":"code","source":"startDay = 0\ntimesteps = 14","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"daysBeforeEventTest = df_days['is_event_day_before'][1941:1969]\ndaysBeforeEvent = df_days['is_event_day_before'][startDay:1941]\ndaysBeforeEvent.index = df_train.iloc[6:,:].index[startDay:1941]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_final = pd.concat([df_train, daysBeforeEvent], axis = 1)\ndf_final.columns","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_final = df_final[startDay:]","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"#Feature Scaling\n#Scale the features using min-max scaler in range 0-1\nfrom sklearn.preprocessing import MinMaxScaler\nscaler = MinMaxScaler(feature_range = (0, 1))\ndt_scaled = scaler.fit_transform(df_final.iloc[6:,:])","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"dt_scaled.shape","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"In the next step, let's create X_train and y_train by creating different dataframes with 14 days of projection. For y_train we only use sales values for predictions. As we only predict sales, only 0:30490 columns are choosen."},{"metadata":{},"cell_type":"markdown","source":"![grafik.png](attachment:grafik.png)","attachments":{"grafik.png":{"image/png":"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"}}},{"metadata":{},"cell_type":"markdown","source":"For each X_train item, 14 past days' sales and 14 daysBeforeEvent feature are included. So one element of X_train's shape is (14, 30491). For y_train we are predicting one day sales of 30490 items therefore one element of y_train's shape is (1, 30490)"},{"metadata":{"trusted":true},"cell_type":"code","source":"X_train = []\ny_train = []\nfor i in range(timesteps, 1941 - startDay):\n    X_train.append(dt_scaled[i-timesteps:i])\n    y_train.append(dt_scaled[i][0:dt_scaled.shape[1]-1]) ","execution_count":null,"outputs":[]},{"metadata":{"scrolled":true,"trusted":true},"cell_type":"code","source":"X_train = np.array(X_train)\ny_train = np.array(y_train)\nprint('Shape of X_train :'+str(X_train.shape))\nprint('Shape of X_train :'+str(y_train.shape))","execution_count":null,"outputs":[]},{"metadata":{"scrolled":true,"trusted":true},"cell_type":"code","source":"gc.collect()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Submission file"},{"metadata":{"scrolled":true,"trusted":true},"cell_type":"code","source":"inputs = df_final[-timesteps:]\ninputs = scaler.transform(inputs)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"df_final[-timesteps:].shape","execution_count":null,"outputs":[]},{"metadata":{"scrolled":true,"trusted":true},"cell_type":"code","source":"gc.collect()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"# Modeling"},{"metadata":{},"cell_type":"markdown","source":"Next, we start our modelling. We will use LSTM Neural Networks with different layers. \n\nIn general, neural networks are easily described by the following picture:\n1. The neural network model starts with random weights and tries to find the best weights for the different layers, predicting outcomes and comparing them with the true target outcomes. For this it uses the loss function. \n2. The loss function measures the quality of  the network’s output\n3. Then, the loss score is used as a feedback signal to adjust the weights."},{"metadata":{},"cell_type":"markdown","source":"![grafik.png](attachment:grafik.png)","attachments":{"grafik.png":{"image/png":"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"}}},{"metadata":{"scrolled":false,"trusted":true},"cell_type":"code","source":"from keras import backend as K","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# defining rmse as loss function\ndef root_mean_squared_error(y_true, y_pred):\n        return K.sqrt(K.mean(K.square(y_pred - y_true)))","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"# defining wrmsse as loss function\n#%run ./WRMSSE.ipynb","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"%who","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"del  df_final, df_train_full, df_days, df_train, dt_scaled, time, sys, signal, reduce_mem_usage","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"gc.collect()","execution_count":null,"outputs":[]},{"metadata":{"scrolled":false,"trusted":true},"cell_type":"code","source":"# Importing the Keras libraries and packages\nimport tensorflow_probability as tfp\nfrom keras.models import Sequential\nfrom keras.layers import Dense\nfrom keras.layers import LSTM\nfrom keras.layers import Dropout\n\n# Initialising the RNN\nmodel = Sequential()\n\n# Adding the first LSTM layer and some Dropout regularisation\nlayer_1_units=40\nmodel.add(LSTM(units = layer_1_units, return_sequences = True, input_shape = (X_train.shape[1], X_train.shape[2])))\nmodel.add(Dropout(0.2))\n\n# Adding a second LSTM layer and some Dropout regularisation\nlayer_2_units=400\nmodel.add(LSTM(units = layer_2_units, return_sequences = True))\nmodel.add(Dropout(0.2))\n\n# Adding a third LSTM layer and some Dropout regularisation\nlayer_3_units=400\nmodel.add(LSTM(units = layer_3_units))\nmodel.add(Dropout(0.2))\n\n# Adding the output layer\nmodel.add(Dense(units = y_train.shape[1]))\n\n# Compiling the RNN\nmodel.compile(optimizer = 'adam', loss = root_mean_squared_error)\n# alternative loss 'mse' or wrmsse","execution_count":null,"outputs":[]},{"metadata":{"scrolled":false,"trusted":true},"cell_type":"code","source":"# Fitting the RNN to the Training set\nepoch_no=100 # going through the dataset 32 times\nbatch_size_RNN=44 # with each training step the model sees 44 examples\nfit = model.fit(X_train, y_train, epochs = epoch_no, batch_size = batch_size_RNN)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"plt.plot(fit.history['loss'])\nplt.title('model loss')\nplt.ylabel('loss')\nplt.xlabel('epoch')\nplt.legend(['train'], loc='upper left')\nplt.show()","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"model.layers","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"gc.collect()","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"## Part a: Modeling with Limited Features"},{"metadata":{"trusted":true},"cell_type":"code","source":"X_test = []\nX_test.append(inputs[0:timesteps])\nX_test = np.array(X_test)\npredictions = []\n\nfor j in range(timesteps,timesteps + 28):\n    predicted_volume = model.predict(X_test[0,j - timesteps:j].reshape(1, timesteps, 30491))\n    testInput = np.column_stack((np.array(predicted_volume), daysBeforeEventTest[0 + j - timesteps]))\n    X_test = np.append(X_test, testInput).reshape(1,j + 1,30491)\n    predicted_volume = scaler.inverse_transform(testInput)[:,0:30490]\n    predictions.append(predicted_volume)","execution_count":null,"outputs":[]},{"metadata":{},"cell_type":"markdown","source":"When creating X_test, we are using the last 14 days in order to predict day 1915 sales.\nTherefore, in order to predict 1916th day, 13 days from our input data and 1 day from our prediction are used. After that we slide the window one by one, i.e.:\n* 12 days from input data + 2 days from our prediction to predict 1917th day\n* 11 days from input data + 3 days from our prediction to predict 1918th day\n* .....\n* 14 days our prediction to predict last 1941th day sales.\n\n"},{"metadata":{},"cell_type":"markdown","source":"![grafik.png](attachment:grafik.png)","attachments":{"grafik.png":{"image/png":"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"}}},{"metadata":{"trusted":true},"cell_type":"code","source":"submission = pd.DataFrame(data=np.array(predictions).reshape(28,30490))\n\nsubmission = submission.T\n    \nsubmission = pd.concat((submission, submission), ignore_index=True)\n\nsample_submission = pd.read_csv('/kaggle/input/m5-forecasting-accuracy/sample_submission.csv')\n    \nidColumn = sample_submission[[\"id\"]]\n    \nsubmission[[\"id\"]] = idColumn  \n\ncols = list(submission.columns)\ncols = cols[-1:] + cols[:-1]\nsubmission = submission[cols]\n\ncolsdeneme = [\"id\"] + [f\"F{i}\" for i in range (1,29)]\n\nsubmission.columns = colsdeneme\n\nsubmission.to_csv(\"submission_evaluation_newloss_nodrop.csv\", index=False)","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"code","source":"predictions","execution_count":null,"outputs":[]},{"metadata":{"trusted":true},"cell_type":"markdown","source":"Please feel free to comment!\nThanks for upvoting!"}],"metadata":{"kernelspec":{"name":"python3","display_name":"Python 3","language":"python"},"language_info":{"name":"python","version":"3.7.6","mimetype":"text/x-python","codemirror_mode":{"name":"ipython","version":3},"pygments_lexer":"ipython3","nbconvert_exporter":"python","file_extension":".py"}},"nbformat":4,"nbformat_minor":4}